Engineering Microbial Rhodopsins as Optical Voltage Sensors Neuroscientists have long dreamed of a genetically encoded sensor that gives an optical signal in response to a change in membrane potential, with the goal of imaging electrical activity of neurons in vivo. Such a molecule could also be used to probe membrane potentials in mitochondria, cardiac cells, bacteria, or in other non-neuronal cells, and thus would provide a new window into the physiological states of a wide range of cells implicated in human health and disease. We propose to engineer a fluorescent transmembrane protein whose fluorescence is sensitive to membrane potential. The goal is to visualize a single action potential in vivo. Many groups have sought to attain this goal; our approach is entirely different from previous efforts. Our starting material is a microbial rhodopsin protein called green proteorhodopsin (GPR). In the wild, this protein absorbs sunlight and pumps protons to generate a proton motive force. We will engineer the protein to run backward-to use membrane voltage to modulate light. The retinal chromophore in wild-type microbial rhodopsins is sufficiently fluorescent for single-cell imaging. GPR can be expressed and imaged in zebra fish neurons in vitro and in living zebra fish. A single-point mutation to GPR leads to a protein whose fluorescence is exquisitely sensitive to membrane potential. The essence of the idea is to use membrane potential to pull a proton toward or away from a color- determining functional group in the protein. When the cell is at rest, this functional group is deprotonated and the protein is dark. When the cell fires an action potential, a proton is forced onto this functional group and the protein becomes bright. Just as GFP revolutionized biology through its ability to track the positions of proteins in cells, we believe that microbial rhodopsins will have a broad impact through their ability to label biological membranes, and to transduce membrane potential into changes in fluorescence.